Pennsylvania is broken. It is an economically and culturally fragmented jurisdiction. Without coherence, the scale of the Commonwealth fails to contribute anything to the prosperity of its people. In the industrial era, railroads tied its urban centres together, creating vast networked economies across the state and beyond.  Today, our two great metropolitan centers struggle alone, and our dozens of smaller cities almost uniformly suffer dire poverty in their cores, isolated from the larger cities and from each other. The transportation function of those steam-era railroads have now been mostly replaced by a large network of state-owned highways. While more versatile than the railroads were, highways encourage prosperity-sapping sprawling land use, and also themselves require enormous and increasing funds to build and maintain. The Pennsylvania Turnpike was America’s First Superhighway, but as we approach its 80th anniversary, the system that has grown from it is failing us. We are constantly pestered into pouring ever more money into a system that makes us poorer.

To address the problem requires three approaches.

The first is to get out of the spiral, and end the building of new highway lane-miles. This should be the strong policy of the Commonwealth, as soon as possible. Very short segments of an interchange or an equivalent of an interchange in size can be permitted to continue, but the trunk routes are done. We’ve built all of the ones that will ever show an economic return, and a few more besides.

The second is to impose congestion-variable tolls on existing highways, wherever it is feasible, with the revenue kept in local regions to maintain political support. Tolls are not a magic bullet, but they do uniquely prevent highways from the usual sprawl cycle of encouraging land use demand around themselves until they become overcapacity as commuter routes. Highways should serve towns, not the other way around.

The third, and the main focus of this post, is to re-leverage the physical assets that highways represent into a transportation option that supports cities and towns instead of dissipating them. We have sunk a lot of capital into those roads, so our goal now has to be maximizing the return we get on that investment.

Link to the map in case the embed fails:,-78.971359,7.33z/data=!4m2!6m1!1s1Ru8nm2OyjKNGdIxYfKDorttf1GEdOCp0

This is a proposal for a full intercity surface transit network for the Commonwealth of Pennsylvania, implementable within 5 years of commencement.  The primary component is a core network of bus services, provided on a fully accessible intercity motorcoach fleet, but the plan also encompasses PennDOT-funded rail service.  The core network bus routes are distinguished from their counterparts today by being far more frequent; at launch, lines marked in purple run hourly or better during the day; lines in green run once every 90-120 minutes.

The collapse this past February of Bieber Tourways blew a large hole in an already fragile intercity travel network in Pennsylvania.  The 72-year-old bus company’s demise brought with it, briefly, the end of intercity service in Reading for the first time in 180 years.  Other intercity bus companies are trying to step in and provide service in the major markets that the end of Bieber has left stranded.  But the intercity bus network today is still sparse, and on top of that is difficult and expensive to use.

Demonstration that the current intercity bus network is unuseable, in the form of a 5 hour bus odyssey as suggested by Google Maps.
This is an actual suggestion from Google Maps for the “fastest” bus directions from Philadelphia to Allentown. Yes, it routes via Atlantic City and Newark.

By vastly increasing the frequency of intercity services, a truly statewide economy can begin to grow again, as businesses and people begin to create networks focused on the smaller cities of Pennsylvania.  Those smaller cities are rich in people and in legacy physical infrastructure.  Many are hosts to institutions of higher learning.  But most are desperately poor in access to opportunity and financial capital.  Very few have hopeful prospects for the future.  By anchoring them together and to the large city economies of Pittsburgh, Philadelphia, and New York, we can give ourselves a chance to thrive together, as one state and one region.



The main trunks are a set of services that run at least once hourly.  The Amtrak Keystone Corridor is one of these, and connects to several other services at Harrisburg, Lancaster, and Philadelphia.  The main bus routes connect Erie and Washington, PA to Pittsburgh; State College to Harrisburg; Harrisburg and York to Baltimore; Reading and the Lehigh Valley to Philadelphia; and each of Northeast Pennsylvania, the Lehigh Valley, and Greater Philadelphia to New York City.

All of these points have fast, Interstate-standard highways (or the Pennsylvania Turnpike) connecting them, allowing for 50 mph average speeds between metropolitan areas.  A large goal of this program is to maximize the return from the sunk costs of these highways, turning them into transit assets that can be used to reinforce the economic strength of our central cities, instead of just sucking away economic activity to the exurban periphery, or being the instrument of destruction for neighborhoods in their path.  It should be the strong policy of the Commonwealth that it has built its last lane-mile of highway trunk, and that future capacity and decongestion needs should be met through transit.  To this end, Turnpike tolls raised by Act 44, which have been the subject of recent controversy, should be used to directly support parallel or hosted transit services, both local and intercity, in applicable markets, and this should be supplemented by additional tolls on existing, currently free, highways.

These bus routes, both the primary trunks and the less-frequent services that feed them, are envisioned as a networked whole.  Some journeys, like York-Philadelphia, involve a transfer.  Where possible, those transfers will be timed, since 60+ minute waits defeat the purpose of having fast intercity transit, but the emphasis will be on ever-increasing frequency of service on those lines.  Branching service to give each possible city-pair a one seat ride is detrimental to the goal of frequency.

What these services are not, is a commute-hour only service.  Where applicable, extra service can exist in the peak commute hours, but this is not envisioned as merely a service for supercommuters.  Business travel, certainly, will be a major market segment, but a morning meeting should not result in a long wait for an evening return journey.  Even among supercommuters, who are often merely dual-wage-earner households with employment (temporarily) split between two urban centers, there must be schedule freedom to admit the possibility of non-9-to-5 schedules.

Scope and Cost

The initial entry-into-service of this network will require, in very approximate figures, 100-150 buses, depending on variables like required spare ratios, and additional peak frequencies or travel time.  Fleet purchase alone will cost $50-120 million.  This starting wave would be spread across no fewer than six, more likely eight or nine, operating bases, each with its own startup capital cost.  Running the network will require a unionized operator headcount of 350-500, a higher ratio than with a local transit agency, since hours-of-service and mandatory rest rules will be more of a concern.  Additional headcount will be required for maintenance, management, and back-end operations.  Doing more comprehensive and precise cost estimates is well beyond my competence.


While the map largely uses existing intercity bus facilities across the Commonwealth for illustrative purposes, this should not be taken as a commitment; existing facilities may or may not be suitable for the increased service foreseen, or they may be poorly-located with respect to their city centers, or conversely they may have poor connectivity to fast arterials or exclusive bus lanes in and out of the city.  Existing facilities have a path-dependent history behind them, and if better options present themselves, then they should be pursued.

In the cases of Philadelphia, Allentown, and New York, the multiple options available can be viewed in a separate map layer.  In the case of Allentown, some Downtown location is needed, but the Wescosville Park and Ride may be a superior transfer location between local Lehigh Valley service and connections along the Northeast Extension to Wilkes-Barre.

One exception to this general rule is facilities that are shared between bus and rail.  In order for connections to be available, and for parallel services to complement each other on services like Pittsburgh-Harrisburg where rail cannot as yet provide enough daily frequencies, all service from a city must be co-located at a single station.

The Norristown Option

High Speed Line car at Norristown Transporation Center, November 2013
The intercity bus terminal at Norristown Transportation Center is partially visible here, on the ground level of the parking garage.

The proposed route network has a heavy reliance on the intercity bus terminal at Norristown Transportation Center as a primary gateway for the Greater Philadelphia area.  This is mainly simple pragmatism: the worst congestion in Pennsylvania is along the Schuylkill Expressway and into Center City Philadelphia.  Cutting out this area from the bus network makes the entire rest of the schedule much more reliable.  Intercity buses may still run to and from Center City, but with that extra layer of isolation, traffic delays east of the Conshohocken Curve cannot badly disrupt a bus journey from Allentown to Wilkes-Barre.  Directing all trips through Norristown also enables much better connections both within the intercity network and to local destinations in the Philadelphia suburbs, which are strong and numerous in their own right.  Noted on the map, but not included in the intercity network as out-of-scope, are proposed regional express bus connections to the Philadelphia International Airport, and to Trenton Station via the Pennsylvania Turnpike.

And of course, Norristown already has a high-capacity option into Center City Philadelphia, in the form of SEPTA Regional Rail.  In the event of fare and staffing reform in which the Manayunk/Norristown Line becomes cheaper and more frequent, it is envisioned that intercity bus tickets have the option of being through-ticketed with a train connection.  Under present circumstances, the high, uncoordinated fare would completely undermine the goal of fare affordability. SEPTA should move to fare reform for its own sake, but the extra bonus of additional guaranteed riders may serve as a financial carrot.  This high level of integration with local and regional carriers is the aspirational goal across the system, but nowhere is it more critical than the Norristown-Philadelphia connection.

The proposed Lafayette Street Interchange with the Pennsylvania Turnpike, aimed at bringing road connectivity to the urban core of Norristown, is probably misguided in its current form, and only escapes the label of a complete boondoggle through two things: the planned extension of the Norristown street grid through underutilized light industrial land to the river’s edge, and the Turnpike’s existing, unusually long spacing between exits.  However, only by adding a strong regional and intercity transit option, and greatly reducing trip times for same, can it become the asset to Norristown that Montgomery County officials want it to be.  (There are other, more difficult problems with the project that need addressing separately.)

A public service, publicly provided

As a technical matter, this system is not reliant on the means by which it is provided.  The failing system we have now, is built on a patchwork of routes provided by many private-sector operators, who set their own fares, rarely co-operate, and only receive a relative pittance of direct subsidy from the Commonwealth.  It is plausible that the entire system could be provided by a series of franchise contracts with private providers, with the Commonwealth setting fares and providing operating support.  For a few routes that barely touch Pennsylvania where the main demand comes from through-passengers, like Cleveland-Erie-Buffalo, this may be the only feasible method of providing state-sponsored service.

However, all of the in-state routes, and a good handful of the interstate routes operating from the main hubs, would instead be better operated by a statewide intercity bus agency set up for the purpose, and reporting directly to PennDOT.  This allows the entire network to internalize its economies of scale, providing better value for money.  It also insulates the services from the vagaries of the national intercity bus market, which is a struggling industry with thin margins, and also the vicissitudes of Harrisburg politics; an administration bent on sabotage can eventually dismantle a public agency over the course of years, even over the objections of committed legislators, but contracts with private providers can be cancelled or allowed to expire at a whim.  And finally, while it is likely that any operating profit that the routes generate will be plowed back into recapitalizing the fleet, adding service, and opening new routes, the principle should be that those funds should be available for those purposes in their totality, and not have a dividend skimmed off the top for the enrichment of distant Scottish shareholders.

At a stroke, this would create one of the top ten, if not top five, transit agencies in the Commonwealth, by fleet size.  Some amount of maintenance and storage co-location by contract with local transit agencies would be possible with a public agency, which would save on costs, but it is not clear whether this would be desirable, or in two cases (Pittsburgh and Norristown), whether it is even feasible at the required scale.

Beyond our borders

We live in a Federal system with full customs and monetary union.  Which is a fancy way of saying, we are surrounded by our fellow Americans and their American dollars, and both are free to come here.  That is the short reason why this state-sponsored network conspicuously does not stop at state borders.  Access to the financial, cultural, educational, and transportation resources of New York City are as critical to the economies of Eastern Pennsylvania, as the connections within the state’s boundaries.  While the political temptation will be to neglect routes out of state, in favor of entirely in-state routes, they serve the needs of the people equally, and should be funded accordingly.  Moreover, our network can serve as an anchor to longer-distance regional connections, especially to and through Ohio and Upstate New York, as provided either by those states or by traditional private-sector bus companies.

Parochialism, a sad inevitability, can be channeled instead into the location of maintenance facilities, driver sign-on locations, fuel and supply contracts, and/or back-office functions.  All of those are potentially lucrative economic assets, much as it feels nauseating to have to re-educate Harrisburg in this particular technique of “honest graft”.

Coverage as well as Ridership

Not every route marked on the map is envisioned as a high-frequency, ridership-oriented express route.  This is a transportation project with an important (secondary) political goal: the knitting together of Pennsylvania’s small cities and towns.  To that end, some routes are envisioned to provide broader coverage, rather than to focus on the core network.  A few of these are even envisioned to be run only once or twice per day in each direction, as demand dictates.  These are referred to on the map as “Lifeline” services.  The flagship of these is the Lifeline service across the Northern Tier, connecting over a dozen small towns with Erie on one end and Scranton on the other, via U.S. Route 6.  This will be, for some towns, their first local passenger service since the 1930s and the twilight of the steam railroad era; in other cases, there is no previous common carrier service recorded at all.  The commercial case for serving these towns on an individual basis is non-existent; the imperative to aggregate them together and connect all of them to the wider Commonwealth, is nearly compulsory.

In this sense, this is the final evolution in the replacement of the marginal steam railroads with highways to connect places, by replicating the railroads’ oldest social contract. If the highway, by being an impediment and a nuisance inside of town, creates value elsewhere for others by enabling fast travel through the town, it must also create value in the town, by connecting it to the world with affordable, accessible, common-carrier passenger transportation.

While the Route 6 corridor makes for an obvious and somewhat ideal route for this kind of service, since all the stops are lined up along one road, I imagine that the final configuration of this class of route will be a political decision, made by political decisionmakers.  As it should be.

The Future On Rails

Something this proposal expressly encompasses is the idea that, if the proposed bus network does its job of stimulating travel and commerce throughout the Commonwealth, then parts of it may be replaced by rail service, as the next logical progression to higher-order capacity.  The obvious places to do this are the short-distance core lines into Philadelphia: Reading and the Lehigh Valley.  While the success of intercity bus makes the case for rail service to Reading and Bethlehem far more compelling, simple restoration of local regional rail is not a 1:1 substitute for express bus service, and rail restoration must be done with those differences in mind.  Very briefly, SEPTA Regional Rail has a much lower average speed because it makes far more stops, so an express service must be overlaid on top of the local stopping service in order to maintain intercity connectivity.  This can be done, but it needs careful scheduling on these constrained lines.

For related reasons, I have long been a skeptic that Regional Rail is a good solution for the intercity markets from Pennsylvania to New York, whether that’s from the Lehigh Valley via the Raritan Valley, or from Scranton via the Lackawanna Cutoff.  The average speeds are just too low to be competitive with the freeways across New Jersey.  However, if NJ Transit is willing to consider a revolution in how it thinks of cross-state transit, either by overlaying express service itself, or by inviting Amtrak or an equivalent provider on to their lines to do it for them, I won’t be one to look that gift horse in the mouth.  I just don’t think it’s very likely to be tried under any circumstances.  This is a different question from whether the Lehigh Valley and Northeastern Pennsylvania could use Regional Rail as transit trunks internal to their own regions, which they can both do very well, if they can get their hands on a legacy railroad and convert it to an electrified, passenger-primary right of way.

The culmination of this progression is the tying together of the east and west of Pennsylvania with greenfield-built (or Turnpike right-of-way-built, where applicable) High Speed Rail, running at 350 km/h or more between Pittsburgh and Harrisburg, and beyond to the Northeast Corridor.  In such a future scenario, which is to say, hopefully very soon after PennDOT decides to stop blatantly sandbagging the project and study actual construction costs from international peers, the Commonwealth-wide bus network will feed into the HSR trunk at multiple points, accelerating its commercial success.  Bringing ever-more economic partners, throughout the Northeast and Midwest, closer and closer to all parts of the Commonwealth, will be not just achievable, but an obvious policy goal within in our grasp.


This is a work in progress, and is in many ways also inextricably linked to other work in progress on future transit options for Pennsylvania, and Southeastern Pennsylvania in particular.  That said, I am eager to hear any feedback and answer any questions prompted by this proposal.  In particular, I am afraid that in some areas, my lack of local knowledge may have created obvious mistakes that others would have caught easily.


Robert Jackel and Alon Levy viewed and submitted feedback to an early draft of this plan, but all of the awkward phrasings, factual errors, and lack of sense, are mine and mine alone.


Join the Conversation


  1. I’d like to see an approximate costing of this network. Otherwise, fantastic piece. Love how you’ve emphasized how the network can help unite the regions of the state together. Right now, it’s city vs rural, east vs west, etc…

  2. I think there’s also something to be said for peak population -> peak tourism. For instance, I would want Philly – Poconos (I actually looked at one point in the past to see if this was possible). Maybe Pittsburgh – Gettysburg? (Bonus prize for fitting in as many Burghs and Burgs as possible along that route.) Unfortunately I don’t think Fallingwater is feasible.

    1. Regional connection from the NEPA cities to the Poconos resort areas is definitely important, but also not something I can speak to well as a mapping thing. (And would probably make the map itself even more illegible.)

      Philly-Gettysburg is there as a two seat ride, Philly-Lancaster (timed rail-bus transfer)-Gettysburg.

Leave a comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s